1. FGF-2 is bound to perlecan in the pericellular matrix of articular cartilage, where it acts as a chondrocyte mechanotransducer 
T.L. Vincent, M.D., Ph.D., C.J. McLean, B.Sc., L.E. Full, B.Sc., D. Peston, B.Sc., J. Saklatvala, Ph.D. 
Osteoarthritis and Cartilage 
Volume 15, Issue 7, Pages (July 2007)
DOI: /j.joca
Copyright © 2007 Osteoarthritis Research Society International Terms and Conditions

2. Fig. 1
Schematic representation of loading unit. The rig was housed in a humidified incubator at 37°C, 20% O2, 5% CO2. One loading unit out of 6 is shown.
Osteoarthritis and Cartilage  , DOI: ( /j.joca )
Copyright © 2007 Osteoarthritis Research Society International Terms and Conditions

3. Fig. 2
Localisation of HSPGs in articular cartilage. Six-micrometre frozen sections of human articular cartilage were pre-treated with hyaluronidase (0.01%, 1h) with or without heparitinase (10mu/ml for 1h). Sections were stained with either the HS stub antibody 3G10 (1:1000) and perlecan (1:1000) (A), with antibodies to perlecan (1:1000) and beta 1 integrin (1:1000) (B), or perlecan (1:1000) and type VI collagen (1:5000) (C). Sections were washed and incubated with secondary FITC labelled antibodies (1:2000). Fluorescence signals were detected by Ultraview confocal microscopy. Bar=20μm.
Osteoarthritis and Cartilage  , DOI: ( /j.joca )
Copyright © 2007 Osteoarthritis Research Society International Terms and Conditions

4. Fig. 3
Immunolocalisation of FGF-2 and perlecan in human and porcine articular cartilage. Articular cartilage from adult human was snap frozen and sectioned (5μm). Sections were treated with hyaluronidase (0.01% for 1h) then stained with an irrelevant antibody (control), or with antibodies to FGF-2 (1:1000) or perlecan (1:1000). Antigens were visualised using the peroxidase based ABC kit (see Methods). Details are shown from the superficial and mid zones of cartilage (A). Bar=100μm. Five-micrometre sections of porcine metacarpophalangeal articular cartilage were treated with hyaluronidase and immunostained as in (A) for FGF-2 and perlecan (B). Six-micrometre sections from human cartilage were taken and incubated with primary antibodies to FGF-2 and perlecan as above. Sections were washed and incubated with secondary FITC labelled antibodies (1:2000). Fluorescence signals were detected either alone (‘Perlecan’ and ‘FGF-2’), or ‘merged’ by Ultraview confocal microscopy (C). Bar=20μm.
Osteoarthritis and Cartilage  , DOI: ( /j.joca )
Copyright © 2007 Osteoarthritis Research Society International Terms and Conditions

5. Fig. 4
Perlecan purified from articular cartilage binds FGF-2 in vitro. FGF-2 was coupled to the IASYS sensor surface (see Methods). 0.8μg (5μl in 4M GnHCl) of perlecan (solid line) or 1.6μg (5μl in 4M GnHCl) of aggrecan (dashed line) was diluted in PBS (95μl) and added to the sensor surface. Binding (arc seconds) was observed (A). In a separate experiment, 0.8μg of perlecan was exchanged into buffer (NaCl 150mmol/CaCl2 5mmol, pH 6.2) then incubated with either buffer alone or 2mu/ml heparitinase for 1h at 37°C. Buffer treated perlecan (solid line), or heparitinase treated perlecan (dashed line) were added to the cuvette in PBS. Binding was observed when each sample (5μl in 95μl PBS) was added to the cuvette sensor surface (B).
Osteoarthritis and Cartilage  , DOI: ( /j.joca )
Copyright © 2007 Osteoarthritis Research Society International Terms and Conditions

6. Fig. 5
Accumulation of pericellular FGF-2 and perlecan in alginate encapsulated chondrocytes. Chondrocytes were digested from porcine articular cartilage and encapsulated into alginate beads (see Methods). Beads were cultured in 10% serum or ITS for 3 weeks. In addition, some freshly isolated chondrocytes were encapsulated into alginate, and cultured for 2 days in serum. At the end of the culture period, beads were dissolved in EDTA, resuspended in a small volume of NaCl (150mM) and then allowed to adhere to Lab-Tek II chamber slides. Cells were immunostained for perlecan (1:1000) or FGF-2 (1:1000) using the peroxidase based ABC kit, as before. Bar=100μm.
Osteoarthritis and Cartilage  , DOI: ( /j.joca )
Copyright © 2007 Osteoarthritis Research Society International Terms and Conditions

7. Fig. 6
An FGF-dependent activation of ERK occurs when alginate encapsulated chondrocytes are loaded. Alginate encapsulated chondrocytes which had been cultured in 10% serum for 3 weeks were made into alginate plugs (see Methods), then serum starved for 2h. Plugs were then left untreated (−) or loaded (+) for 2min (0.01MPa, 0.5Hz). After 30min chondrocytes were dissolved from the alginate (50mM EDTA in 150mM NaCl with 2% Na3VO4) on ice and lysates generated to immunoblot for phosphorylated ERK. The blot was stripped and reprobed for total ERK (A, Encapsulated chondrocytes). Similar experiments were performed with porcine knee articular cartilage explants loaded for 2min (0.13MPa, 0.5Hz) (A, explants), and with alginate encapsulated chondrocytes grown for 3 weeks in ITS (B). To test the FGF-2 dependence of ERK activation upon loading, alginate encapsulated chondrocytes which had been cultured for 4 weeks in 10% serum were pre-incubated with either medium alone (control) or SB (250nM) for 48h. Plugs were prepared as above, and were treated with either serum-free medium (c), IL-1 (100ng/ml), FGF-2 (100ng/ml) or were loaded cyclically for 2min (0.01MPa, 0.5Hz) (load). After 30min, lysates were generated as above for phosphorylated ERK and total ERK immunoblots (C). Rested cartilage explants were cyclically loaded with 0.032, 0.01 and 0.002MPa for 2min. Explants were lysed as above and immunoblotted for phosphorylated, then total ERK (D).
Osteoarthritis and Cartilage  , DOI: ( /j.joca )
Copyright © 2007 Osteoarthritis Research Society International Terms and Conditions

8. Fig. 7
ERK activation upon loading is dependent upon pericellular FGF-2. Alginate encapsulated chondrocytes were generated in three different ways: cells which had been freshly isolated and encapsulated for 2 days only (control), cells which had been encapsulated and cultured in 10% serum for 3 weeks (10% serum), and cells which had been encapsulated and cultured in 10% serum for 3 weeks but in whom additional recombinant FGF-2 had been added to the culture for 3 days in week 2 (10% serum/FGF-2). At the end of the culture period some cells were dissolved from alginate and immunostained for FGF-2 as before (A). The remainder of cells were made in to alginate plugs as before. Plugs were either left untreated (C), stimulated with FGF-2 (100ng/ml), or were loaded for 2min (0.01MPa, 0.5Hz). Cell lysates were immunoblotted for phosphorylated and total ERK as before (B).
Osteoarthritis and Cartilage  , DOI: ( /j.joca )
Copyright © 2007 Osteoarthritis Research Society International Terms and Conditions

9. Fig. 8
Protein tyrosine phosphorylation upon chondrocyte loading is mainly FGF-2 dependent. Chondrocytes were extracted from articular cartilage, encapsulated into alginate beads, and cultured in 10% serum for 4 weeks. Forty-eight hours prior to loading some beads were incubated with SB (250nM). Beads were made into plugs on the day of the experiment as described above. Plugs were treated with either serum-free medium (C), IL-1 (100ng/ml), FGF-2 (100ng/ml), or were loaded (0.01MPa, 0.5Hz). After 15min plugs were dissolved with EDTA on ice and lysed. Lysates were run on 12.5% sodium dodecyl sulphate page, then immunoblotted with an anti-phosphotyrosine antibody (1:3000) (A). The blots were stripped and reprobed first for phosphorylated ERK (B), then total ERK (C).
Osteoarthritis and Cartilage  , DOI: ( /j.joca )
Copyright © 2007 Osteoarthritis Research Society International Terms and Conditions